An FDG-PET-Based Machine Learning Framework to Support Neurologic Decision-Making in Alzheimer Disease and Related Disorders.

IF 7.7 1区 医学 Q1 CLINICAL NEUROLOGY
Neurology Pub Date : 2025-07-22 Epub Date: 2025-06-27 DOI:10.1212/WNL.0000000000213831
Leland Barnard, Hugo Botha, Nick Corriveau-Lecavalier, Jonathan Graff-Radford, Ellen Dicks, Venkatsampath Gogineni, Gemeng Zhang, Brian J Burkett, Derek R Johnson, Sean J Huls, Aditya Khurana, John L Stricker, Hoon-Ki Paul Min, Matthew L Senjem, Winnie Z Fan, Heather Wiste, Mary M Machulda, Melissa E Murray, Dennis W Dickson, Aivi T Nguyen, R Ross Reichard, Jeffrey L Gunter, Christopher G Schwarz, Kejal Kantarci, Jennifer L Whitwell, Keith Anthony Josephs, David S Knopman, Bradley F Boeve, Ronald C Petersen, Clifford R Jack, Val J Lowe, David T Jones
{"title":"An FDG-PET-Based Machine Learning Framework to Support Neurologic Decision-Making in Alzheimer Disease and Related Disorders.","authors":"Leland Barnard, Hugo Botha, Nick Corriveau-Lecavalier, Jonathan Graff-Radford, Ellen Dicks, Venkatsampath Gogineni, Gemeng Zhang, Brian J Burkett, Derek R Johnson, Sean J Huls, Aditya Khurana, John L Stricker, Hoon-Ki Paul Min, Matthew L Senjem, Winnie Z Fan, Heather Wiste, Mary M Machulda, Melissa E Murray, Dennis W Dickson, Aivi T Nguyen, R Ross Reichard, Jeffrey L Gunter, Christopher G Schwarz, Kejal Kantarci, Jennifer L Whitwell, Keith Anthony Josephs, David S Knopman, Bradley F Boeve, Ronald C Petersen, Clifford R Jack, Val J Lowe, David T Jones","doi":"10.1212/WNL.0000000000213831","DOIUrl":null,"url":null,"abstract":"<p><strong>Background and objectives: </strong>Distinguishing neurodegenerative diseases is a challenging task requiring neurologic expertise. Clinical decision support systems (CDSSs) powered by machine learning (ML) and artificial intelligence can assist with complex diagnostic tasks by augmenting user capabilities, but workflow integration poses many challenges. We propose that a modeling framework based on fluorodeoxyglucose PET (FDG-PET) imaging can address these challenges and form the basis of an effective CDSS for neurodegenerative disease.</p><p><strong>Methods: </strong>This retrospective study focused on FDG-PET images in a discovery cohort drawn from 3 research studies plus routine clinical patients. When selecting research study participants, the inclusion criterion was the availability of an FDG-PET image from within 2.5 years of diagnosis with 1 of 9 specific neurodegenerative syndromes or designation as unimpaired. Participants from disease groups were recruited from the clinical patient population while unimpaired participants came primarily from a population study. The discovery cohort was used to develop a clinical decision support framework we call StateViewer, which applies a neighbor matching algorithm to detect the presence of 9 different neurodegenerative phenotypes. The ML performance of this framework was evaluated in the discovery cohort by nested cross-validation and externally validated in the Alzheimer's Disease Neuroimaging Initiative. Potential for clinical integration was demonstrated in a radiologic reader study focused on differentiating posterior cortical atrophy from Lewy body dementia.</p><p><strong>Results: </strong>The discovery cohort contained 3,671 individuals with a mean age of 68 years and consisted of 49% reported female. Our model framework was able to detect the presence of 9 different neurodegenerative phenotypes with a sensitivity of 0.89 ± 0.03 and an area under the receiver operating characteristic curve of 0.93 ± 0.02. In the radiologic reader study, readers using our model were found to have 3.3 ± 1.1 times greater odds of making a correct diagnosis than readers using a current standard-of-care workflow.</p><p><strong>Discussion: </strong>Our proposed framework provides strong classification performance with high interpretability, and it addresses many of the challenges that face clinical integration of ML-based decision support tools. One limitation of this study is a uniform discovery cohort that is not representative of other patient populations in some regards.</p>","PeriodicalId":19256,"journal":{"name":"Neurology","volume":"105 2","pages":"e213831"},"PeriodicalIF":7.7000,"publicationDate":"2025-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12207676/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Neurology","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1212/WNL.0000000000213831","RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2025/6/27 0:00:00","PubModel":"Epub","JCR":"Q1","JCRName":"CLINICAL NEUROLOGY","Score":null,"Total":0}
引用次数: 0

Abstract

Background and objectives: Distinguishing neurodegenerative diseases is a challenging task requiring neurologic expertise. Clinical decision support systems (CDSSs) powered by machine learning (ML) and artificial intelligence can assist with complex diagnostic tasks by augmenting user capabilities, but workflow integration poses many challenges. We propose that a modeling framework based on fluorodeoxyglucose PET (FDG-PET) imaging can address these challenges and form the basis of an effective CDSS for neurodegenerative disease.

Methods: This retrospective study focused on FDG-PET images in a discovery cohort drawn from 3 research studies plus routine clinical patients. When selecting research study participants, the inclusion criterion was the availability of an FDG-PET image from within 2.5 years of diagnosis with 1 of 9 specific neurodegenerative syndromes or designation as unimpaired. Participants from disease groups were recruited from the clinical patient population while unimpaired participants came primarily from a population study. The discovery cohort was used to develop a clinical decision support framework we call StateViewer, which applies a neighbor matching algorithm to detect the presence of 9 different neurodegenerative phenotypes. The ML performance of this framework was evaluated in the discovery cohort by nested cross-validation and externally validated in the Alzheimer's Disease Neuroimaging Initiative. Potential for clinical integration was demonstrated in a radiologic reader study focused on differentiating posterior cortical atrophy from Lewy body dementia.

Results: The discovery cohort contained 3,671 individuals with a mean age of 68 years and consisted of 49% reported female. Our model framework was able to detect the presence of 9 different neurodegenerative phenotypes with a sensitivity of 0.89 ± 0.03 and an area under the receiver operating characteristic curve of 0.93 ± 0.02. In the radiologic reader study, readers using our model were found to have 3.3 ± 1.1 times greater odds of making a correct diagnosis than readers using a current standard-of-care workflow.

Discussion: Our proposed framework provides strong classification performance with high interpretability, and it addresses many of the challenges that face clinical integration of ML-based decision support tools. One limitation of this study is a uniform discovery cohort that is not representative of other patient populations in some regards.

基于fdg - pet的机器学习框架支持阿尔茨海默病和相关疾病的神经学决策。
背景和目的:鉴别神经退行性疾病是一项具有挑战性的任务,需要神经学专业知识。由机器学习(ML)和人工智能驱动的临床决策支持系统(cdss)可以通过增强用户能力来协助完成复杂的诊断任务,但工作流集成带来了许多挑战。我们提出基于氟脱氧葡萄糖PET (FDG-PET)成像的建模框架可以解决这些挑战,并为神经退行性疾病的有效CDSS奠定基础。方法:本回顾性研究集中于从3项研究和常规临床患者中抽取的发现队列中的FDG-PET图像。在选择研究参与者时,纳入标准是诊断为9种特定神经退行性综合征之一或指定为未受损的2.5年内的FDG-PET图像的可用性。来自疾病组的参与者是从临床患者人群中招募的,而未受损的参与者主要来自人群研究。发现队列用于开发我们称为statviewer的临床决策支持框架,该框架应用邻居匹配算法来检测9种不同神经退行性表型的存在。该框架的ML性能在发现队列中通过嵌套交叉验证进行评估,并在阿尔茨海默病神经影像学倡议中进行外部验证。临床整合的潜力在一项放射学研究中得到证实,该研究的重点是区分后皮层萎缩和路易体痴呆。结果:发现队列包含3671人,平均年龄为68岁,其中49%为女性。我们的模型框架能够检测到9种不同神经退行性表型的存在,灵敏度为0.89±0.03,接受者工作特征曲线下面积为0.93±0.02。在放射学阅读器研究中,使用我们模型的阅读器比使用当前标准护理工作流程的阅读器做出正确诊断的几率高3.3±1.1倍。讨论:我们提出的框架提供了强大的分类性能和高可解释性,并且它解决了基于ml的决策支持工具的临床集成所面临的许多挑战。本研究的一个局限性是发现队列是统一的,在某些方面不能代表其他患者群体。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
Neurology
Neurology 医学-临床神经学
CiteScore
12.20
自引率
4.00%
发文量
1973
审稿时长
2-3 weeks
期刊介绍: Neurology, the official journal of the American Academy of Neurology, aspires to be the premier peer-reviewed journal for clinical neurology research. Its mission is to publish exceptional peer-reviewed original research articles, editorials, and reviews to improve patient care, education, clinical research, and professionalism in neurology. As the leading clinical neurology journal worldwide, Neurology targets physicians specializing in nervous system diseases and conditions. It aims to advance the field by presenting new basic and clinical research that influences neurological practice. The journal is a leading source of cutting-edge, peer-reviewed information for the neurology community worldwide. Editorial content includes Research, Clinical/Scientific Notes, Views, Historical Neurology, NeuroImages, Humanities, Letters, and position papers from the American Academy of Neurology. The online version is considered the definitive version, encompassing all available content. Neurology is indexed in prestigious databases such as MEDLINE/PubMed, Embase, Scopus, Biological Abstracts®, PsycINFO®, Current Contents®, Web of Science®, CrossRef, and Google Scholar.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信